NextFin News - Deutsche Bank Research is leaning against the idea that artificial intelligence will deliver a broad productivity payoff any time soon. The bank’s macro team is arguing that the technology’s most important effects are still working their way through adoption, investment and workflow change, while the economy-wide gains investors want to see will take time to show up in the numbers.
That view sits at the center of a larger market debate. AI has already become one of the dominant investment themes in global equities, but the question that matters for economists is different from the question that matters for traders. Investors are focused on the companies selling chips, cloud capacity and software tools. Economists are asking when those purchases will translate into higher output per hour across the wider economy.
Deutsche Bank Research has said the technology could unlock particularly large productivity gains in Germany, where many areas of the economy remain relatively underdigitized. The same research argues that the current period is marked by persistent AI-driven optimism alongside geopolitical disruption, a combination that makes the macro outlook unusually hard to read.
In that setting, the message is not that AI is irrelevant. It is that the lag between adoption and measurable productivity improvement is likely to be long enough to disappoint anyone expecting a quick macro boost. For now, the strongest evidence is still showing up in spending plans, strategic positioning and competitive pressure rather than in a clean, economy-wide productivity break.
Adoption Can Be Fast; Productivity Usually Is Not
The key distinction is between buying the tools and absorbing the gains. A company can add AI software, data infrastructure and automated workflows quickly. It can also tell investors that margins should improve. But the broader economy only records a productivity benefit once those tools are deeply embedded in production, administration and service delivery.
That is why the timetable matters. New technologies often move through a long adjustment period in which early adopters get the first benefits and late adopters see little change. Training workers, rewriting processes, replacing legacy systems and integrating new software into everyday operations all take time. Until those changes are widespread, official productivity statistics can remain stubbornly unimpressive even if management teams are enthusiastic.
Deutsche Bank’s own research points to a similar mechanism in Germany. If large parts of the economy are still only partially digitized, then the potential productivity upside from AI may be larger, but the adjustment process may also be longer and more uneven. In other words, the bigger the possible gain, the more work it may take to realize it.
That is especially relevant in economies facing labor shortages and aging populations. AI may help companies cope with missing workers or bottlenecks before it visibly raises productivity growth at the national level. The benefit is real, but it does not necessarily appear first in the broad statistics that investors tend to watch.
“The world economy is grappling with a complex interplay of persistent AI-driven optimism and the disruptive force of the Middle East conflict, making it feel like 1999 meets 1990, but hopefully not 1973.”
That line captures the broader backdrop. The AI story is unfolding in a world still shaped by geopolitics, supply constraints and uneven growth. Even when the technology is powerful, the path from adoption to measurable macro impact can be slow, uneven and easy to overstate in the short run.
What The Debate Really Means For Markets
The market problem is not whether AI will matter. It already does, because firms are spending on the infrastructure and software needed to compete in the new environment. The harder question is whether that spending will show up soon enough in the aggregate data to justify the speed of the enthusiasm.
If AI gains remain concentrated in a handful of companies and sectors, then the macro numbers may continue to look lagging even as investors keep rewarding the perceived winners. That would leave the AI trade intact, but it would also mean that the trade is being driven more by relative advantage than by a broad productivity renaissance.
By contrast, a stronger macro case would require the gains to spread across industries and into the national accounts. That would mean more than a few headline examples. It would require sustained evidence of shorter workflows, lower error rates, faster product cycles and better utilization of labor and capital across a wide range of businesses.
For now, the more cautious interpretation has the better evidence behind it. The technology is clearly moving fast at the company level, but the economy-wide payoff is still a second-order question. That is why Reid’s skepticism lands with force: the AI story may be real, but the productivity story is still developing.
What comes next is a slower test. Corporate earnings, capex plans, hiring decisions and official productivity releases will show whether AI is changing firm behavior faster than it is changing the broader economy. If the gains spread, the current caution will look too conservative. If they stay narrow, the market will have to keep justifying its optimism with revenue and margin gains rather than with a macro boom that has not yet arrived.
The bottom line is simple. AI may already be reshaping the investment landscape, but productivity gains are still a diffusion story, not a headline statistic. Markets can reprice the future quickly; economies usually need longer.
Explore more exclusive insights at nextfin.ai.
